作 者: (倪小威); (刘迪仁); (艾林); (冯加明); (梁霄); (敖旋峰);
机构地区: 油气资源与勘探技术教育部重点实验室(长江大学),武汉430100 长江大学地球物理与石油资源学院,武汉430100
出 处: 《煤炭技术》 2017年第9期105-107,共3页
摘 要: 从煤层气储层测井响应机理出发,优选出煤储层及顶、底板产水敏感测井参数。利用BP神经网络算法对平均日产水量进行预测,准确率在76%左右,并具备区域产水预测效果。 Based on logging response mechanism of CBM reservoir, the sensitive well logging parameters of water production of CBM reservoir and roof and floor bed are optimized. The BP neural network algorithm is used to forecast the average daily water production. The accuracy rate is about 76%, and the method has the effect of regional water production forecasting.